It has been hypothesized that the pathogenesis of diseases induced by cigarette smoking involves oxidative damage by free radicals. However, definitive evidence that smoking causes the oxidative modification of target molecules in vivo is lacking. We conducted a study to determine whether the production of F2-isoprostanes, which are novel products of lipid peroxidation, is enhanced in persons who smoke. We measured the levels of free F2-isoprostanes in plasma, the levels of F2-isoprostanes esterified to plasma lipids, and the urinary excretion of metabolites of F2-isoprostanes in 10 smokers and 10 nonsmokers matched for age and sex. The short-term effects of smoking (three cigarettes smoked over 30 minutes) and the effects of two weeks of abstinence from smoking on levels of F2-isoprostanes in the circulation were also determined in the smokers. Plasma levels of free and esterified F2-isoprostanes were significantly higher in the smokers (242 +/- 147 and 574 +/- 217 pmol per liter, respectively) than in the nonsmokers (103 +/- 19 and 345 +/- 65 pmol per liter; P = 0.02 for free F2-isoprostanes and P = 0.03 for esterified F2-isoprostanes). Smoking had no short-term effects on the circulating levels of F2-isoprostanes. However, the levels of free and esterified F2-isoprostanes fell significantly after two weeks of abstinence from smoking (250 +/- 156 and 624 +/- 214 pmol per liter, respectively, before the cessation of smoking, as compared with 156 +/- 67 and 469 +/- 108 pmol per liter after two weeks' cessation; P = 0.03 for free F2-isoprostanes and P = 0.02 for esterified F2-isoprostanes). The increased levels of F2-isoprostanes in the circulation of persons who smoke support the hypothesis that smoking can cause the oxidative modification of important biologic molecules in vivo.